This specification relates to speech recognition.
A user of a mobile device may enter text by, for example, typing on a keyboard or speaking into a microphone. In the context of voice input, an Automated Search Recognition (ASR) engine may be used to process input in the form of speech.